Anomaly Detection and Reporting for Machine Learning Models

9. The system of claim 1, wherein the report includes a first indication of the input data anomaly, a second indication of the output data anomaly, and a third indication of the contribution of the input data anomaly to the output data anomaly.

14. The method of claim 12, wherein the detecting the output anomaly comprises determining that a difference between values of one or more performance metrics and historical values of the one or more performance metrics is greater than a threshold amount based at least in part on comparing the values to the historical values.

16. The method of claim 12, wherein the report includes an indication of a type of anomaly for each of the first input anomaly and the second input anomaly, the type of anomaly corresponding to at least one of an input data quality issue or an input data distribution issue.

20. The computer system of claim 19, wherein the report includes an indication of a type of anomaly of the input anomaly, the type of anomaly corresponding to at least one of an input data quality issue or an input data distribution issue.